# Why do many people prefer roll-high to roll-under?

#### mamba

##### Legend
You need to do a slight mathematical translation to figure out the odds for roll-high.

If your target number is 15, on roll under the odds are just the number times 5%, assuming that matching the number is a success.

If your target number is 15, on roll over the odds are (21-target) times 5%, assuming that matching the number is a success.
yes, but that is trivial enough, and once you add bonuses the roll under ‘falls apart’, either you have to subtract the bonus from the roll (addition is easier than subtraction) or you have to add it to the value you need to stay under rather than the roll, and that is a little weird.

So all in all I consider that a wash

#### billd91

##### Not your screen monkey (he/him) 🇺🇦🇵🇸🏳️‍⚧️
For me, "Roll Under" systems make it massively more difficult to establish the difficulty of the task at hand as something that's objectively part of the "game world" and not a mechanical artifact.

It's the big problem I always had with GURPS. It never felt like doing anything that happened was based on the situation my character was in, and more about, what does the number on my character sheet say the situation is.

It was an incredibly strange disconnect for me, having this supposedly ultra-realistic, highly granular "game world model" engine, yet the most basic of all mechanics -- a skill check -- is abstracted in a way that always felt weird and artificial, from the first time I played GURPS until the last time I played it after ~25 sessions.

It was so jarring and terrible that I'll never really consider a "roll under" system ever again.
I wouldn't be surprised if the primary disconnect comes from rolling multiple dice under a value that doesn't map to an easily intuitable value. Ever play Call of Cthulhu? The percentile values always struck me as very easy to interpret since an Anthropology score of 70% means 70% chance of success on the die roll and it translates to being knowledgeable about 70% of the field of anthropology (for the setting time period - By Gaslight, 1920s, Now).

#### Staffan

##### Legend
Do players just associate bigger numbers with being "better" in some way? Is there some cultural factor I'm missing? Is this just a way of saying "I don't want to play any game that isn't like D&D"?
Yes, more big number more better. It also feels like it's more easy to obscure difficulty without off-loading work on the players. "I rolled a 23, do I succeed" is easier on the player than "I rolled 5 below my skill, do I succeed?" (because the latter involves the player calculating margin of success), so there's more of an assumption that the GM will tell you modifiers up front in roll-under systems.

Of course, roll-under systems have plenty of advantages too, but for those who do prefer roll-high, I think those two are the main ones.

#### innerdude

##### Legend
I wouldn't be surprised if the primary disconnect comes from rolling multiple dice under a value that doesn't map to an easily intuitable value. Ever play Call of Cthulhu? The percentile values always struck me as very easy to interpret since an Anthropology score of 70% means 70% chance of success on the die roll and it translates to being knowledgeable about 70% of the field of anthropology (for the setting time period - By Gaslight, 1920s, Now).

But even then, if I as GM want to modify the difficulty, it's counter-intuitive.

"I want to penalize this roll by 20%, because it's 20% more difficult than normal."

Counter-intuitively, you either have to subtract 20% from the target, or raise the roll by 20%.

If I want to make something harder, it's naturally more intuitive to raise the target number, not reduce it.

Likewise, as a player, it's counter-intuitive to say, "I'm penalizing you by 20%, so you have a +20 penalty."

But the bigger deal to me is still that roll-under naturally obscures the idea of objective difficulty. Roll under always orients the player to view the difficulty as represented on their character sheet, not as an unknown factor in the game world itself. I realize that mathematically it's a complete wash, it makes no difference in actual calculation. But in my head it makes a huge difference in how I think about task resolution.

#### schneeland

##### Hero
I agree that subtraction is slightly more effort than addition, but I find numeric comparison easier than addition. I may be unusual in this: numbers are more strongly and obviously ordered for me than letters of the alphabet.
I would agree that comparison is faster in general, but number ranges also matter. At least personally, I feel that addition and comparison are both relatively fast for single digit numbers and a moderate amount of dice (I would place the limit lower for addition than comparison - probably 3 vs. 10 dice).
However, if we compare a typical d100 system to a 2d6+X system, then both might be about equally fast (or slow), since the number range for the first one is higher.

#### schneeland

##### Hero
Counter-intuitively, you either have to subtract 20% from the target, or raise the roll by 20%.
I agree with your general assessment, but one option I find more intuitive is to use two thresholds in such cases, i.e. difficulty as the lower bound and skill as the upper bound (e.g. Whitehack does something like this for AC in combat).

#### TheAlkaizer

##### Game Designer
Personally, I'm fine with either.

But my players are always taken aback the first time I tell them their goal is to roll as low as possible. Their character is getting better, some of the numbers on their sheet are going up, but their still rolling the same die hoping to get as low as possible. From their perspective, the only thing that changes is the treshold, or tolerance to how low you need to roll to succeed.

In a roll above, you often make the use of bonuses which players, in my experience, just tend to catch more quickly. You want to roll as high as possible, the die is randomness and your bonus is how good you are at it. Your character gets better, so does your bonus. You improve while the challenges of the world stay on the level that they are.

It's mostly a matter of perception in my experience.

#### John Dallman

##### Hero
It's the big problem I always had with GURPS. It never felt like doing anything that happened was based on the situation my character was in, and more about, what does the number on my character sheet say the situation is.
Did your GM not use situational modifiers? That's how you reflect the difficulty of a situation in GURPS, just as you set DCs in d20 systems.

#### Staffan

##### Legend
But even then, if I as GM want to modify the difficulty, it's counter-intuitive.

"I want to penalize this roll by 20%, because it's 20% more difficult than normal."

Counter-intuitively, you either have to subtract 20% from the target, or raise the roll by 20%.

If I want to make something harder, it's naturally more intuitive to raise the target number, not reduce it.

Likewise, as a player, it's counter-intuitive to say, "I'm penalizing you by 20%, so you have a +20 penalty."

But the bigger deal to me is still that roll-under naturally obscures the idea of objective difficulty. Roll under always orients the player to view the difficulty as represented on their character sheet, not as an unknown factor in the game world itself. I realize that mathematically it's a complete wash, it makes no difference in actual calculation. But in my head it makes a huge difference in how I think about task resolution.
The Troubleshooters has a neat way of doing difficulty in a roll-under system. If something is easy, you roll at "+N pips", and if it's hard you roll at "-N pips". If you're rolling at +N pips, any ones roll of 1 through N is a success, regardless of your skill, and if you're rolling at -N pips any ones roll of 1 through N is a failure regardless of skill (so if you have skill 69 and are rolling at +2 pips, you would succeed at 01-69, 71-72, 81-82, and 91-92).

#### Epic Meepo

I would assume roll-over is generally more popular because many people are conditioned by a variety of societal factors to assume the larger of two numbers is better.

That being said, I'm going to cast my vote for Blackjack- or Price-is-Right-style mechanics, where you try to role as high as possible without going over a certain limiting factor (which you also want to be as high as possible).

Replies
31
Views
6K
Replies
14
Views
5K
Replies
215
Views
18K
Replies
50
Views
2K
Replies
4
Views
912